Guides
Named Entity Recognition

Named Entity Recognition

Build a Named Entity Recognition (NER) model to extract structured information from text.

What You’ll Build

An AI model that identifies and classifies named entities in text. Common use cases:

  • Extract names, dates, locations from documents
  • Identify products and prices in descriptions
  • Parse medical records for conditions and treatments
  • Extract legal entities from contracts

Prerequisites

  • A SeeMe.ai account (sign up)
  • Text documents for training
  • (Optional) Python environment with seeme SDK installed

Example

Input:

“Apple Inc. was founded by Steve Jobs in Cupertino, California on April 1, 1976.”

Output:

EntityType
Apple Inc.ORGANIZATION
Steve JobsPERSON
Cupertino, CaliforniaLOCATION
April 1, 1976DATE

Steps

1. Prepare Data 2. Annotate Entities 3. Train Model 4. Deploy 5. Monitor

Supported Formats

  • spaCy format (JSON with token offsets)
  • IOB/BIO format (token-level labels)
  • Custom JSON format

SeeMe.ai uses spaCy under the hood for NER training.